INDEX

Explanations

names of philosophers and political figures

oai_token-act-pair · gpt-3.5-turbo

references to prominent philosophers and thinkers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kodi

-0.78

LOAD

-0.75

 Nationals

-0.71

hap

-0.71

Lago

-0.69

SHARE

-0.68

ittee

-0.67

unity

-0.66

cue

-0.64

 installer

-0.64

POSITIVE LOGITS

 philosopher

0.95

zsche

0.91

ocrates

0.90

ian

0.90

 scholar

0.90

 scholars

0.88

 relativity

0.86

 philosophers

0.86

icist

0.85

olkien

0.85

Activations Density 0.245%

names of philosophers and political figures

references to prominent philosophers and thinkers

No Comments

No Known Activations